Available for research collaborations

Liang
GUO

Researcher in Renewable Energy Systems · Power Electronics · AI-based Optimization

I develop AI-driven energy management strategies for hybrid clean energy systems — from hydrogen-powered maritime vessels to fuel cell hybrid vehicles. My work sits at the intersection of reinforcement learning, power electronics, and real-time hardware validation.

01 — About

Bridging AI and Clean Energy Systems

I am a researcher specializing in intelligent energy management for hybrid renewable energy systems. My research develops reinforcement learning algorithms — including Deep RL, Fuzzy-RL hybrids, and policy gradient methods — to optimize energy flows in fuel cell, battery, and hydrogen-powered systems.

During my PhD at Aix-Marseille University, I pioneered Fuzzy-RL approaches for fuel cell hybrid electric vehicles, combining interpretable fuzzy logic with adaptive reinforcement learning for real-time energy dispatch. My work was validated through Hardware-in-the-Loop (HIL) platforms using dSPACE, OPAL-RT, and Typhoon-HIL.

Currently at ENSM, I lead energy management research for the Navire Propre project, developing hydrogen–diesel–battery hybrid propulsion systems for low-carbon maritime vessels. Beyond AI-driven energy management, I focus on shipboard power system modeling and real-time simulation, with Power Hardware-in-the-Loop (PHIL) validation using Imperix platforms.

416+
Citations
10
H-Index
16
Publications
9+
Years HIL Exp.
02 — Research Experience

Current & Past Research

2024 — Present
Researcher — Hydrogen Maritime Hybrid Propulsion
Ecole Nationale Superieure Maritime (ENSM) · France
Leading energy management strategy design for the Navire Propre project — retrofitting and optimizing hydrogen‑diesel‑battery hybrid propulsion systems for low-carbon ships. Work spans multi-objective system sizing, shipboard DC/AC microgrid modeling, real-time simulation, and PHIL validation with Imperix platforms.
Hydrogen Fuel CellHybrid PropulsionShipboard MicrogridDigital TwinImperix PHILAI EMS
2024
Postdoctoral Researcher — Hybrid Energy Storage Systems
Aix-Marseille University · LIS UMR CNRS 7020 · France
EU Horizon Project FlexCHESS: developed flexibility services for grid-connected and islanded Hybrid Energy Storage Systems (HESS). Conducted optimal sizing, real-time simulation, and HIL validation for multi-objective HESS configurations.
HESSFlexibility ServicesReal-Time SimOptimal Sizing
2020 — 2023
PhD — AI-driven Energy Management for FCHEV
Aix-Marseille University · France · PECASE & LIS Labs
Developed Fuzzy-RL hybrid algorithms (Fuzzy REINFORCE, Fuzzy Q-Learning, Fuzzy SARSA, Fuzzy Double Q-Learning) for real-time energy management in fuel cell hybrid electric vehicles. Validated on dSPACE SCALEXIO and OPAL-RT HIL platforms with physical fuel cell and battery hardware.
Reinforcement LearningFuzzy LogicFCHEVHILFuel Cell Lifetime
2017 — 2020
Master Researcher — Power Electronics & Robust Control
Northwestern Polytechnical University · China
Designed high-order sliding mode controllers for bidirectional DC/DC converters with SiC/GaN wide-bandgap devices. Built HIL platforms for fuel cell UAV power systems. Contributed to NSFC-funded fuel cell UAV endurance optimization project.
Sliding Mode ControlSiC/GaN ConvertersDC/DCUAV Power
03 — Publications

Selected Works

2026
Conf.
PEMD Global 2026 · London, UK Accepted
Adaptive droop control for shipboard fuel cell hybrid DC microgrids with voltage restoration and SOC regulation.
2025
Conf.
IEEE ISGT Europe 2025 · Valletta, Malta DOI: 10.1109/isgteurope64741.2025.11305404
On-policy Fuzzy SARSA applied to fuel cell-powered maritime propulsion energy management.
2023
Journal
Energy and AI, 2023 IF: 9.6
Combines fuzzy membership functions with policy gradient RL for interpretable, high-performance energy management in FCHEVs.
Conf.
IEEE ISIE 2023 · Helsinki-Espoo, Finland DOI: 10.1109/isie51358.2023.10227988
Solves Q-value overestimation bias by integrating fuzzy membership with double Q-learning.
2022
Conf.
IEEE ICSC 2022 · Marseille, France DOI: 10.1109/icsc57768.2022.9993916
Fuel cell lifetime optimization via self-learning fuzzy Q-learning with degradation model integration.
Conf.
IEEE IECON 2022 · Brussels, Belgium DOI: 10.1109/iecon49645.2022.9968966
Rule-based fuzzy value function combined with reinforcement learning for FCHEV energy dispatch.
Conf.
ICEAI 2022 · Belfort, France Best Presentation Award
Policy gradient method with fuzzy membership functions for interpretable RL-based energy management.
2021
Conf.
IEEE IECON 2021 · Toronto, Canada DOI: 10.1109/iecon48115.2021.9589725 · Top Conference
Foundational RL-based energy management framework for fuel cell hybrid electric vehicles.
2019
Journal
IEEE Trans. on Transportation Electrification, 2019 IF: 6.5 · 68 citations · DOI: 10.1109/tte.2019.2943895
High-order sliding mode control with noise suppression for bidirectional converters in fuel cell electric vehicles.
Journal
IEEE Trans. on Power Electronics, 2019 IF: 7.65 · 62 citations · DOI: 10.1109/tpel.2019.2905897
ADR voltage control of floating interleaved DC-DC boost converter with switch fault tolerance.
Conf.
IEEE IAS 2019 · Baltimore, USA DOI: 10.1109/ias.2019.8912436
HOSM observer with tanh smoothing for fuel cell UAV power system uncertainty estimation.
Conf.
IEEE ICIT 2019 · Melbourne, Australia DOI: 10.1109/icit.2019.8843697
Smooth SMC with chattering suppression for high-conversion-ratio bidirectional converters.
2018
Conf.
IEEE PEAC 2018 · Shenzhen, China DOI: 10.1109/peac.2018.8590428
MPC with Kalman compensation for interleaved DC-DC boost converter current control.
2017
Journal
IEEE Trans. on Industrial Electronics, 2017 IF: 8.162 · 107 citations · DOI: 10.1109/tie.2017.2703663
Active stabilization of on-board DC power converter systems with input filter interactions.
Conf.
IEEE IECON 2017 · Beijing, China
TS fuzzy model-based virtual capacitor control for cascaded DC power system stability.
2016
Journal
Proceedings of the CSEE, 2016
High-frequency current feed-forward compensation for stability enhancement in boost converter systems.
04 — Skills

Technical Expertise

🤖
AI & Machine Learning
Reinforcement LearningDeep LearningPyTorchFuzzy-RL HybridPolicy GradientQ-Learning
Power Electronics
DC/DC ConvertersSiC/GaN DevicesBidirectional ConverterSliding Mode ControlStability Analysis
🔬
HIL & Real-Time
dSPACE SCALEXIOOPAL-RTTyphoon-HILImperix B-BoxStarSim9+ Years Exp.
🛠
Embedded Systems
STM32TI DSP283xxXilinx ZYNQRaspberry PiJetson NanoArduino/ESP32
💻
Programming
PythonC/C++MATLAB/SimulinkLabVIEWVerilog HDLShell
📡
Protocols & Tools
TCP/IPMQTTModbusCANUART/SPI/I2CAltium Designer
05 — Education

Academic Background

2020 — 2023
PhD in Automation
Aix-Marseille University · France
Full Scholarship · Best Presentation Award ICEAI 2022
“Reinforcement Learning based Energy Management for Fuel Cell Hybrid Electric Vehicles”
2017 — 2020
MSc in Electrical Engineering
Northwestern Polytechnical University · China
Rank 4/248 · National Scholarship (Top 3%)
“Strong Robust Control and Chattering Suppression of Bi-Directional DC/DC Converter”
2013 — 2017
BSc in EE & Automation
Northwestern Polytechnical University · China
Rank 1/94 · Outstanding Graduate · Excellent Thesis
“High-order Sliding Mode Control based Bidirectional DC/DC Converter with SiC MOSFET”
06 — Contact

Let’s Connect

I am open to research collaborations, academic exchanges, and consulting opportunities in renewable energy systems, AI-based optimization, and maritime decarbonization.